I interned on the Encarta team for one summer a few years before its death, had a blast working there, and used Encarta a lot when I was a kid (remember Mindmaze?) so I’ve given this question a bit of thought over the years:
- Can’t just blame Wikipedia: It’s easy to tout Wikipedia’s awesomeness, but it took Wikipedia years to grow to where it is today. Encarta had a trusted dominant brand and had plenty of time to cement that, so I don’t think all the credit for the win goes to Wikipedia.
- Encarta wasn’t Microsoft’s focus anymore: There’s a book “The Microsoft Way” that describes in great detail the really innovative formation of the Encarta product and business. Encarta really helped Windows be successful by being the killer app to get people to upgrade their PCs to have CD-ROM drives and speakers so they could play Encarta’s multimedia content. This is why the team mattered within the company. By Wikipedia’s early founding days in 2000 the shift to multimedia was complete, so Encarta no longer played this key role within Microsoft and gradually became less of a company focus. (There’s a similar answer to why Internet Explorer lost.)
- Unnatural organizations: By the 2000s, Microsoft wasn’t quite sure what to do with any of their desktop products like Encarta, Money, Works, Picture It, and Streets & Trips. Though these products were profitable, their revenue was a drop in the bucket compared to Windows and Office revenues. They used to be important to drive Windows installs, but by this point, people were buying Windows whether these products existed or not. Since they weren’t important to Windows, they needed a new organizational home and most of them got thrown into the MSN division. One unfortunate side effect is this meant they started getting subjected to MSN success metrics. For example, there might be a MSN org goal of increasing ad revenue by x%, and Encarta had to help make that happen, which really made no sense for these desktop programs and resulted in strange product changes. Philip Su, who worked on Microsoft Money, wrote a great piece explaining this dysfunction (http://blogs.msdn.com/b/philipsu/archive/2004/07/01/170682.aspx). Encarta actually had a solid website, but you couldn’t see the good stuff unless you paid $20/month for MSN to be your ISP, and no one I knew used MSN as their ISP. Even then you had to use the MSN Explorer web browser (http://en.wikipedia.org/wiki/MSN_Explorer) to see it. The free Encarta website had a small subset of the content, was much uglier, missing features, was laden with banner ads, and used Passport for login. When being part of MSN didn’t go great, Microsoft next tried tying the Encarta team to Office, resulting in the team getting distracted building products like “Learning Essentials for Microsoft Office” and “Microsoft Math” to drive Office sales.
- Hard to become a web company: Though Encarta.com launched and was a moderate success, it’s taken a while for Microsoft to actually shift to web. Only recently with Bing have people started to think of Microsoft as a first class web company, which is well after Encarta’s death. It was a challenge to retool an engineering team that was focused on a Windows desktop product to be great at building a website. It’s not just a technology/programming challenge; successful web teams operate very differently from successful desktop teams. The much faster release cycle is one difference that immediately comes to mind. I also don’t think it’s a stretch for me to say that it was hard for them to recruit the best talent in the world, or even from within the company, to work on MSN, let alone on encarta.msn.com. Building on ASP/VBScript and IIS certainly didn’t help them recruit experienced web talent externally, and this started during the dot com boom when talent was scarce.
- Innovator’s dilemma: One of the reasons Encarta was initially successful is that the content was great. They came to fame with videos but eventually it permeated all the content in the CD product. Great quality was important because a lot of Encarta sales were people buying the product for their kids. Quality was ensured because literally half the team was non-technical editors that vetted everything. Editorial was a first class function/role (http://www.seattlepi.com/business/article/Microsoft-Notebook-Encyclopedia-editor-finds-his-1149114.php). Competing with Wikipedia’s breadth of content meant embracing the community though, which guaranteed that there were windows of time where the content was inaccurate or even obscene. The editorial team found it unacceptable to tarnish the brand in this way, and since they always had an equal vote at the table, the team never fully embraced the community model. They thought they could continue to win by focusing on quality content. Eventually they half-heartedly let users propose article edits, which wouldn’t get committed until an editor approved them, but that really was insufficient to build a community that cared about contributing. Plus this came years too late. This is standard innovator’s dilemma where something with a lower cost structure that starts at laughably bad quality slowly gets better and eventually eats your lunch. It would have taken the Encarta team guts to significantly trim down the editorial staff and compromise content quality, and that never happened.
Answer by Wayne Kao